Text

Introduction: Juvenile Idiopathic Arthritis (JIA) comprises diverse chronic inflammatory conditions driven by malfunction of the immune system. The intestinal microbiota is considered a crucial environmental factor correlating with chronic inflammatory diseases. In this study, we have characterized the JIA microbiota taxonomically and phenotypically in relation to age dynamics and therapy response to methotrexate (MTX).

Methods: We have analyzed the intestinal microbiota of 54 JIA patients and 38 age-matched healthy controls using 16S rRNA sequencing and multi-parameter microbiota flow cytometry (mMFC). With mMFC, we interrogate coating of bacteria with endogenous host immunoglobulins by isotype-specific stainings and the surface expression of specific sugar moieties using sugar-specific plant-derived lectins. The multivariate data are analyzed with machine learning algorithms to define disease- or therapy-specific microbial signatures.

Results: We found that age is a major confounder in the microbiota analyses of children. When stratifying patients and healthy controls according to age into three groups, we could show that each age group has a distinct, non-overlapping taxonomic and phenotypic signature (PERMANOVA, all p < 0.05). Specifically, we identified novel bacterial species associated with JIA not seen in the non-stratified comparison between JIA patients and healthy controls. Furthermore, we also identified age group-specific phenotypic signatures differing between JIA patients and healthy controls based on immunoglobulin coating and lectin staining. Using mMFC and based on microbial signatures, we were able to discriminate responders and non-responders to MTX-therapy (PERMANOVA, R2= 0.28, p=9.9*10-5) and also to predict therapy responsiveness (AUC = 0.873, 95 % CI: 0.691 – 1).

Conclusion: Our results highlight the dynamic nature of the intestinal microbiota in pediatric patients and show the necessity to account for age in microbiome analyses in children. The mMFC approach shown here, demonstrates that single-cell analysis of the intestinal microbiota from stool samples has the potential to impact not only therapy monitoring but also therapy prediction, a first step towards microbiota-guided personalization of medicine.

Disclosures: The authors declare that they have no competing interests.